Issue in retriving the PM Mass Concentration values of Sensiron sen66

[KaviGowtham1]

Hello,

I have interfaced my ESP32 with the Sensirion SEN66 sensor. While using the embedded I2C Sen66 driver you provided, I observed that the PM mass concentration values are being received incorrectly, whereas all other parameters are received correctly. I have attached the relevant proofs for reference. Could you please clarify this issue as soon as possible?
I've Attached the code i've written, the Output of it and the PM original Value confirmed by the SensorBridge (control center)

Here is the code i've written:

`#include <stdio.h>
#include <stdint.h>
#include <stdbool.h>

#include "sensirion_config.h"
#include "sensirion_common.h"
#include "sensirion_i2c.h"
#include "sensirion_i2c_hal.h"
#include "sen66_i2c.h"

#define SEN66_I2C_ADDR SEN66_I2C_ADDR_6B

void app_main(void) {
int16_t err;
uint8_t padding = 0;
bool data_ready = false;

// Init I2C HAL (GPIO 21/22 in sensirion_i2c_hal.c)
sensirion_i2c_hal_init();
sensirion_i2c_hal_select_bus(0);
sen66_init(SEN66_I2C_ADDR);

printf("SEN66 demo starting...\r\n");

err = sen66_start_fan_cleaning();
if (err != NO_ERROR) {
    printf("Error fan cleaning: %d\r\n", err);
    return;
}

err = sen66_start_continuous_measurement();
if (err != NO_ERROR) {
    printf("Error start: %d\r\n", err);
    return;
}

sensirion_i2c_hal_sleep_usec(1500000); // 1.5 s warm-up
while (1) {
    uint16_t mc_pm1p0, mc_pm2p5, mc_pm4p0, mc_pm10p0;
    int16_t  rh_raw, t_raw, voc_index, nox_index;
    uint16_t co2_ppm;	
    err = sen66_get_data_ready(&padding, &data_ready);
    if (err != NO_ERROR) {
        printf("DataReady err: %d\r\n", err);
        break;
    }
    if (!data_ready) {
        sensirion_i2c_hal_sleep_usec(200000);
        continue;
    }

    // Mass concentration + T/RH/VOC/NOx/CO2
    err = sen66_read_measured_values_as_integers(
        &mc_pm1p0, &mc_pm2p5, &mc_pm4p0, &mc_pm10p0,
        &rh_raw, &t_raw, &voc_index, &nox_index, &co2_ppm);
    if (err != NO_ERROR) {
        printf("Read mass conc err: %d\r\n", err);
        break;
    }
		
	// PM MASS concentration
	float pm1p0  = mc_pm1p0  / 10.0f;
	float pm2p5  = mc_pm2p5  / 10.0f;
	float pm4p0  = mc_pm4p0  / 10.0f;
	float pm10p0 = mc_pm10p0 / 10.0f;
	
	// RH, Temperature, VOC, NOx 
	float rh     = rh_raw    / 100.0f;
	float temp_c = t_raw     / 200.0f;
	float nox    = nox_index / 10.0f;
	float voc    = voc_index / 10.0f;


    printf("T = %.2f C, RH = %.2f %%\r\n", temp_c, rh);
    printf("Mass: PM1.0=%.1f PM2.5=%.1f PM4.0=%.1f PM10=%.1f ug/m3\r\n",
           pm1p0, pm2p5, pm4p0, pm10p0);
    printf("NOx index=%.1f, CO2=%u ppm, VOC index=%.1f\r\n\r\n",
           nox, (unsigned)co2_ppm, voc);

    sensirion_i2c_hal_sleep_usec(1000000); // 1 s
}


sen66_stop_measurement();
sensirion_i2c_hal_free();

}`

[quentin-cy]

Hello @KaviGowtham1

Thank you for reaching out, I will try to reproduce you error once I have a moment but it indeed seems like something is off.
I hope to reach back this week.

Cheers
Q

[quentin-cy]

Hej @KaviGowtham1

After further investigation, we mostly noticed the following issue: Our ESP32 embedded driver was outdated and hard to actually use.
The #15 adds a ready to run ESP32 dev-kit example (using IDF toolchain).

We also tested on RaspberryPi but observed no such issue.
Using the example above (ESP32), we could observe a couple of readings out of the ordinary but it seemed like it was mostly caused by environment (desk dust behind raised while setting things up)

Could you test with our sample code and make sure to let the environment settle for a while? (with smallest possible code changes)
This way we could rule out those environment factors

Q.

PS: An engineer, also pointed out that the power delivery could be insufficient and causing bad readings. Could you clarify your setup (board, wiring, power delivery)